Unlock the Potential of Machine Learning to Get Better Cost Estimates in the Early Stage of the Product Lifecycle
Machine learning is no longer science fiction – it is real and many businesses are already taking advantage. As a new breed of software that is able to learn without being explicitly programmed, machine learning can access, analyze, and find patterns in data in a way that is beyond human capabilities.
Although machine learning is not a new concept, it has recently gained fresh momentum and promises high potential to get better cost estimates in the early stage of the product lifecycle and in the quotation stage of unique, complex products.
Typically manufacturing companies are confronted with long sales cycles, complex cost structures and supply chains, rising material and transportation cost. At the early stage of the quotation process, in particular for engineer-to-order products, companies leverage templates and reuse other quotes to streamline the quotation process. The cost engineers are looking for similar products and would benefit from a recommender system which produces a list of recommendations, and identifies abnormalities and provide suggestions for the costing structure and the pricec. Moreover, these recommendations are also relevant for product cost estimates in new product development projects e.g. to find similar materials or assemblies for the newly engineered product.
The cost engineers could take full advantage of advanced machine learning capabilities that help gaining smart costing process for greater speed and cost accuracy.
Designed with Manufacturers for Manufacturers
SAP Product Lifecycle Costing is the result of SAP’s collaboration with over 30 manufacturing customers. Having a continuous interaction with co-innovation customers enabled us to create a product costing solution that meets all the key requirements of manufacturers. And we are still adapting the solution to make it even more useful.
The recent co-innovation topic concerns SAP Leonardo Machine Learning, the exciting new offering that extends the analytics and Machine Learning portfolio of SAP. With embedded functional services, intelligence can be added easily to SAP Product Lifecycle Costing solution. That let the cost engineer learn from data and extract knowledge that was previously inaccessible.
Apply machine learning techniques to optimize decisions
Jointly in collaboration with our co-innovation customers we identified several use cases for SAP Product Lifecycle Costing, which could make cost engineers more efficient with machine learning.
Similar Parts and Processes. Find similar or alternative parts to facilitate re-use or get rough price estimate.
Validation and Plausibility Checks. Check abnormal prices and designs, missing or inconsistent entries.
Price and Resource Forecast. Predict prices and other resources for future products and time periods based on previous products or web sources.
Profit and Impact Optimization. Propose improvement measures for production location, supply chain optimization.
SAP Product Lifecycle Costing is boosting efficiency with intelligent search and machine learning for similar parts and processes
The intelligent search and machine learning function is a prototype and was developed in close collaboration with Daimler AG, one of the biggest producers of premium cars and the world’s biggest manufacturer of commercial vehicles with a global reach.
To define the costs of the future product in the early phase of development is not easy, since a lot of materials and processes have to be considered that often still not exist.
Thus, it’s hard to find the appropriate material and price. The cost engineer would like to use materials that is used in other products, for example to conclude from old product model on new models, or to estimate the new variants of models in a bottom-up way based on components and modules. However, this is not an easy task.
How to find the appropriate candidate among thousands of potential similar parts? This is time-consuming.
Here is where machine learning comes into the play. It supports the intelligent search for similar parts according to selected attributes. Machine learning optimizes the search parameters, at the same time still allowing for easy manual fine-tuning of the search function.
Discover how intelligent search and machine learning embedded in SAP Product Lifecycle Costing boost the efficiency.
Watch the video.
More efficient and accurate retrieval of similar items supports quicker and more precise estimation of costs in the classic costing scenarios.
Marion Heidenreich is working in the SAP’s Industrial Machinery & Components Industry Business Unit with industry focus on Finance, Procurement, and business networks. With almost 20 years of SAP experience, she is well versed in business development, solution management and go-to-market for discrete manufacturing industries.